The project will develop estimates of the trends in fertility and mortality of the black population of the U.S. between 1820 and 1950 and produce age distributions corrected for age misreporting and underenumeration for each sex at each census date. The estimates will be derived using new techniques for comparing reported age distributions of mortality with reported age distributions of the population, a new four-parameter model of mortality and a detailed study of mortality differentials at various dates between states in the Death Registration Area (DRA) and non-DRA states. In addiltion computerization of data from a 1% sample of black households in the 1910 census from Maryland, Kentucky, Atlanta and New Orleans will make it possible to estimate the completeness of death registration among blacks. Preliminary estimates suggests that the mortality and fertility rates of blacks in 1880-1900 were substantially lower than previously estimated. It is likely that the estimates derived from the proposed research will require substantial changes in the explanations of the declines in fertility and mortality among blacks. This project will also provide valuable verification of several new methodologies designed for use in developing countries. The availability of extensive data on U.S. blacks make it possible to test these techniques more thoroughly than is possible elesewhere.